NaproxenInducesCell-CycleArrestandApoptosisinHuman …...NaproxenInducesCell-CycleArrestandApoptosisinHuman Urinary Bladder Cancer Cell Lines and Chemically Induced Cancers by Targeting

Research Article

Naproxen Induces Cell-Cycle Arrest and Apoptosis in HumanUrinary Bladder Cancer Cell Lines and Chemically InducedCancers by Targeting PI3K

Mi-Sung Kim1, Jong-Eun Kim1, Do Young Lim1, Zunnan Huang1, Hanyong Chen1, Alyssa Langfald1,Ronald A. Lubet2, Clinton J. Grubbs3, Zigang Dong1, and Ann M. Bode1

AbstractNaproxen [(S)-6-methoxy-a-methyl-2-naphthaleneacetic acid] is a potent nonsteroidal anti-inflam-

matory drug that inhibits both COX-1 and COX-2 and is widely used as an over-the-counter medication.

Naproxen exhibits analgesic, antipyretic, and anti-inflammatory activities. Naproxen, as well as other

nonsteroidal anti-inflammatory drug, has been reported to be effective in the prevention of urinary

bladder cancer in rodents. However, potential targets other than the COX isozymes have not been

reported. We examined potential additional targets in urinary bladder cancer cells and in rat bladder

cancers. Computer kinase profiling results suggested that phosphoinositide 3-kinase (PI3K) is a

potential target for naproxen. In vitro kinase assay data revealed that naproxen interacts with PI3K

and inhibits its kinase activity. Pull-down binding assay data confirmed that PI3K directly binds with

naproxen in vitro and ex vivo. Western blot data showed that naproxen decreased phosphorylation of Akt,

and subsequently decreased Akt signaling in UM-UC-5 and UM-UC-14 urinary bladder cancer cells.

Furthermore, naproxen suppressed anchorage-independent cell growth and decreased cell viability by

targeting PI3K in both cell lines. Naproxen caused an accumulation of cells at the G1 phase mediated

through cyclin-dependent kinase 4, cyclin D1, and p21. Moreover, naproxen induced significant

apoptosis, accompanied with increased levels of cleaved caspase-3, caspase-7, and PARP in both cell

types. Naproxen-induced cell death was mainly because of apoptosis in which a prominent down-

regulation of Bcl-2 and upregulation of Bax were involved. Naproxen also caused apoptosis and

inhibited Akt phosphorylation in rat urinary bladder cancers induced by N-butyl-N-(4-hydroxybu-

tyl)-nitrosamine. Cancer Prev Res; 7(2); 236–45. �2013 AACR.

IntroductionBecause of the great cost and time involved in the

development of new drugs, the use of drug repositioningis particularly appealing. Drug repositioning is defined asthe development of existing drugs for new uses (1).Existing drugs on the market have already been testedfor toxicity and safety and have favorable or validatedpharmacokinetic properties for licensing from the Food

and Drug Administration. Furthermore, development of anew drug usually requires 10 to 17 years, whereas drugrepositioning can decrease the time needed to bring adrug to the patient. Therefore, drug repositioning couldobviously save substantial money and time comparedwith new drug development (2).

Naproxen (Fig. 1A), a potent COX-1 and COX-2 inhib-itor, is one of the best known and commonly used nonste-roidal anti-inflammatory drugs (NSAID). It was introducedin prescription form as "naprosyn" in 1976. Naproxen hasbeen used to treat inflammation associated with conditionssuch as arthritis, ankylosing spondylitis, tendinitis, bursitis,or gout (3). COX-2 selective and nonselective NSAIDs havebeen linked to increases in the risk and number of seriouscardiovascular events. However, naproxen reportedly hasthe least risk of adverse cardiovascular events comparedwith other NSAIDs based on a wide range of epidemiologicstudies (4, 5). Although NSAIDs are regarded as cancerpreventive agents, naproxen might have additional targetmolecules beyond its obviousCOX inhibition,whichmightmake it a more effective cancer preventive agent.

Previous studies have shown that naproxen could inhibitdevelopment of urinary bladder cancer in rodents (6, 7).

Authors' Affiliations: 1The Hormel Institute, University of Minnesota,Minnesota; 2Division of Cancer Prevention, National Cancer Institute,National Institutes of Health; and 3University of Alabama at Birmingham,Birmingham, Alabama

M.-S. Kim and J.-E. Kim contributed equally to this work.

Current address for Z. Huang: China-America Cancer Research Institute,Guangdong Medical College, Guangdong Provincial Key Laboratory ofMedical Molecular Diagnostics, Dongguan, Guangdong, People's Repub-lic of China.

Corresponding Author: Ann M. Bode, The Hormel Institute University ofMinnesota, 801 16thAvenueNE, Austin,MN55912. Phone: 507-437-9615;Fax: 507-437-9606; E-mail: [email protected]

doi: 10.1158/1940-6207.CAPR-13-0288

�2013 American Association for Cancer Research.

CancerPreventionResearch

Cancer Prev Res; 7(2) February 2014236

Research. on May 5, 2020. © 2014 American Association for Cancercancerpreventionresearch.aacrjournals.org Downloaded from

Published OnlineFirst December 10, 2013; DOI: 10.1158/1940-6207.CAPR-13-0288

http://cancerpreventionresearch.aacrjournals.org/

Furthermore, the NSAID, piroxicam, has been proven effec-tive in reducing bladder cancer in dogs (8–10) and at leastsome epidemiologic studies in humans have shown signif-icant preventive efficacy with NSAIDs (11). However, theprecise mechanism and variety of potential direct target(s)of naproxen in urinary bladder cancer prevention have notbeen elucidated.In this study, we used virtual screening to identify

potential novel protein targets of naproxen and identifiedphosphoinositide 3-kinase (PI3K) as a novel target fornaproxen. Naproxen showed anticancer effects against

bladder cancer cells by inducing cell-cycle arrest and apo-ptosis. Naproxen also caused apoptosis and inhibited Aktphosphorylation in rat urinary bladder cancers inducedby N-butyl-N-(4-hydroxybutyl)-nitrosamine (OH-BBN),a urinary bladder carcinogen.

Materials and MethodsReagents

Naproxen was purchased from Cayman Chemical. Dul-becco’s Modified Eagle Medium and other supplements

Figure 1. Naproxen decreases PI3K kinase activity. A, chemical structure of naproxen [(S)-6-methoxy-a-methyl-2-naphthaleneacetic acid]. B, naproxeninhibits PI3K activity in vitro. Active PI3K (100 ng) was mixed with naproxen (0, 0.5, 1, or 2 mmol/L) or LY294002 (i.e., a PI3K inhibitor, 10 mmol/L) and thenincubated with a [g-32P] ATP mixture. The 32P-labeled PIP3 was separated by thin-layer chromatography and then visualized by autoradiography. Alldata are shownasmeans�SDof values from3 independent experiments.Banddensitywasmeasuredusing the ImageJ (NIH) softwareprogram. Theasterisk(�) indicates a significant difference between PI3K kinase activities as determined by t test (P < 0.05). C, naproxen binds to PI3K in vitro. Active PI3K (500 ng)was incubated with naproxen-conjugated EAH Sepharose 4B beads or EAH Sepharose 4B beads alone, and the pulled down proteins were analyzedby Western blot analysis. D, naproxen binds to PI3K ex vivo. Cell lysates from UM-UC-5 bladder cancer cells (500 mg) were incubated with naproxen-conjugated EAH Sepharose 4B beads or EAH Sepharose 4B beads alone, and the pulled down proteins were analyzed by Western blot analysis. Data arerepresentative of 3 independent experiments that gave similar results. E, docking model of naproxen and the PI3K protein structure. Naproxen isshown in sphere representation and carbons are colored white. PI3K is shown as a cartoon model. F, binding site of PI3K with naproxen. The ATP-bindingsite of PI3K is shown in surface representation.

Naproxen Targets PI3K to Prevent Urinary Bladder Cancer

www.aacrjournals.org Cancer Prev Res; 7(2) February 2014 237




were from Life Technologies, Inc. EAH Sepharose 4B beadswere purchased from GE Healthcare Biosciences. Thehuman recombinant PI3K (p110a/p85a) was from Milli-pore Corp. The antibodies against phosphorylated Akt(Ser473), total Akt, phosphorylated mTOR, total mTOR,phosphorylated p70S6K, total p70S6K, cyclin D1, cyclin A,cyclin-dependent kinase (CDK)4, CDK2, cleaved caspase-3(Asp175), cleaved caspase-7 (Asp198), and cleaved PARP(Asp214) were purchased from Cell Signaling Biotechnol-ogy. Antibodies used to detect Bcl-2, Bax, and b-actin werefrom Santa Cruz Biotechnology. The protein assay kit wasfrom Bio-Rad.

Virtual screeningThe coordinates of naproxen were obtained from the

crystal structure of 2VDB (12), which was downloadedfrom the Protein Data Bank (PDB). The cocrystallizedligand was selected as the query molecule structure forshape screening. This database is directly used for shape-similarity screening without any further modification.Shape similarity screening was performed using the phasemodule (13) in the Schr€odinger Suite by comparing thevolumetric similarity between naproxen and each com-pound from the PDB ligand database. During the screeningprocess, polar hydrogens were included and only onealignment per ligand was recorded. In addition, the outputstructures were sorted by decreasing ShapeSim score andthen a ranked list was created. Only the kinases with aShapeSim score of at least 0.75 were selected (Table 1).

Cell cultureUM-UC-5 (gefitinib sensitive) and UM-UC-14 (gefitinib

resistant) human urinary bladder cancer cell lines wereobtained from American Type Culture Collection. The cellswere cultured at 37�Cwith 5%CO2 inDulbecco’s ModifiedEagle Medium (DMEM) supplemented with 10% FBS, 2mmol/L L-glutamine, and 100 units/mL penicillin. Cells

were cytogenetically tested and authenticated before beingfrozen. Each vial of frozen cells were thawed and main-tained for about 2 months (16 passages).

In vivo treatment with naproxenOH-BBN was used to induce urinary bladder tumors as

previously described (14). When a control OH-BBN–treated rat developed a palpable bladder tumor, the ratwas treated with naproxen (400 mg/kg in diet) for 5 days.At the end of this time, animals were sacrificed and thebladder tumors were quickly removed and divided so thatroughly half of each tumor was snap frozen in liquidnitrogen and the other half was fixed for immunohis-tochemistry or immunofluorescence. This presurgicalapproach to examining biomarkers has previously beenused by us to examine the effects of gefitinib on thesetypes of tumors (15).

In vitro PI3K kinase assayThe in vitro PI3K kinase assay was performed as described

previously (16). Active PI3K (100 ng) was incubated withnaproxen (0, 0.5, 1, or 2mmol/L) or LY294002 (10mmol/L)for 10 minutes at 30�C. LY294002, a well-known PI3Kinhibitor, was used as a positive control. The mixtures wereincubated with 0.5 mg/mL phosphatidylinositol (MP Bio-medicals) for 5 minutes at room temperature, followed byincubation with reaction buffer [10 mmol/L Tris-HCl (pH7.6), 60 mmol/L MgCl2, and 0.25 mmol/L ATP containing10 mCi [g-32P] ATP] for an additional 10 minutes at 30�C.The reaction was stopped by adding 15 mL of 4 N HCl and130 mL of chloroform:methanol ¼ 1:1. After mixing, thelower chloroform phase was spotted onto a silica gel plate(Merck KGaA). The resulting 32P-labeled phosphatidylino-sitol-3,4,5-trisphosphate (PIP3) was separated by thin layerchromatography with developing solvent (chloroform:methanol:NH4OH:H2O ¼ 60:47:2:11.3) and then visual-ized by autoradiography.

Table 1. Potential kinase target of naproxen

Ligand Similarity scoring PDB ID Protein

RFZ 0.823 3H30 Human casein kinase 2 (CK2)EMO 0.817 3C13 Human casein kinase 2 (CK2)39Z 0.815 2PE0 Human phosphoinositide-dependent kinase-1 (PDK1)41A 0.812 3DPD Multi-isoform PI3KMYC 0.753 1e90 PI3KV25 0.806 2VAG Di-phosphorylated human CDC2-like kinase 1 (CLK1)GVN 0.779 2uw5 PKA-PKB CHIMERA (PKB)GVI 0.756 2uvy PKA-PKB CHIMERA (PKB)L15 0.753 2uw4 PKA-PKB CHIMERA (PKB)73Q 0.773 2g01 JNK1537 0.760 1uki JNK1675 0.768 1owe Urokinase745 0.751 1u6q UrokinaseABJ 0.759 3en7 c-Src kinase domainB49 0.754 3g0e KIT kinase domain

Kim et al.

Cancer Prev Res; 7(2) February 2014 Cancer Prevention Research238




In vitro and ex vivo pull-down assaysNaproxen-conjugated EAH (1,6-diaminohexane) Sephar-

ose 4B beads were prepared following the instructionsprovided by GE Healthcare Biosciences. For in vitro or exvivo pull-down assays, active PI3K (500 ng) or lysates fromUM-UC-5 cells (500mg)weremixedwith 1mLof naproxen-conjugated EAH Sepharose 4B beads in reaction buffer [50mmol/L Tris-HCl (pH 7.5), 5 mmol/L EDTA, 150 mmol/LNaCl, 1 mmol/L dithiothreitol (DTT), 0.01% NP-40, 2mg/mL bovine serum albumin, 0.02 mmol/L phenyl-methylsulfonylfluoride (PMSF), and protease inhibitorcocktail]. After incubation with gentle rocking at 4�Covernight, the beads were washed 5 times with washingbuffer [50 mmol/L Tris-HCl (pH 7.5), 5 mmol/L EDTA,150mmol/LNaCl, 1mmol/LDTT, 0.01%NP-40, and 0.02mmol/L PMSF]. The proteins bound to the beads wereanalyzed by Western blotting.

Cell viability assayCells were seeded (1 � 103 cells/well) in 96-well plates

with 10% FBS/DMEM and incubated at 37�C in a 5% CO2

incubator for 24 hours and then treatedwith different doses(0, 0.5, 1, or 2 mmol/L) of naproxen. After incubation, 20mL of CellTiter96 Aqueous One Solution (Promega) wereadded to each well. Cells were then incubated for 1 hour at37�C in a 5% CO2 incubator. Absorbance was measured at490 and 690 nm.

Anchorage-independent cell growthCells (8� 103 cells/well) suspended in complete growth

medium (DMEM or Basal Medium Eagle [BME] supple-mented with 10% FBS and 1% antibiotics) were added to0.3% agar with different doses (0, 0.5, 1, or 2 mmol/L) ofnaproxen in a top layer over a base layer of 0.6% agar withdifferent doses (0, 0.5, 1, or 2 mmol/L) of naproxen. Theculturesweremaintained at 37�C in a 5%CO2 incubator for2 weeks and then colonies were counted under a micro-scope using the Image-Pro Plus software (v.6.2) program(MediaCybernetics). All experimentswere repeated 3 times.

Cell-cycle assayCells were seeded (2 � 105 cells/well) in 6-well plates

with 10% FBS/DMEM and incubated overnight at 37�C in a5%CO2 incubator. Cells were then incubated in serum-freemedium for 24 hours followed by treatment for 48 hourswith naproxen (0, 0.5, 1, and 2 mmol/L) in 10% FBS/DMEM. The cells were trypsinized, then washed twice withcold PBS, and finally fixed with ice-cold 70% ethanol at20�C overnight. Cells were then washed twice with PBS,incubated with 20 mg/mL RNase A and 200 mg/mL pro-pidium iodide inPBSat roomtemperature for 30minutes inthe dark, and subjected to flow cytometry using the FACS-Calibur flow cytometer. Data were analyzed using ModFitLT (Verity Software House, Inc.).

Apoptosis assayAnnexin V and propidium iodide staining were used to

visualize apoptotic cells in a similar procedure as described

above but without prefixing with 70% ethanol. Cells werestained using the Annexin V-FITC Apoptosis Detection Kit(MBL International Corporation) and propidium iodideaccording to the manufacturer’s instructions. Cells wereanalyzed by 2-color flow cytometry. The emission fluores-cence of the Annexin V conjugatewas detected and recordedthrough a 530/30 bandpass filter in the FL1 detector.Propidium iodide was detected in the FL2 detector througha 585/42 bandpass filter. Apoptotic cells were only thoselocated in the bottom right quadrant that stained positivefor Annexin V and negative for propidium iodide.

Western blot analysisCell lysates were prepared with radioimmunoprecipita-

tion assay buffer [20 mmol/L Tris-HCl (pH 7.5), 150mmol/L NaCl, 1 mmol/L Na2EDTA, 1 mmol/L EGTA,1% Triton, 2.5 mmol/L sodium pyrophosphate, 1 mmol/Lb-glycerophosphate, 1mmol/LNa3VO4, 1mg/mL leupeptin].Equal amounts of protein were determined using the RCDCprotein assay (Bio-Rad). Proteins were separated by SDS-PAGE and electrophoretically transferred to polyvinylidenedifluoride membranes (GE Healthcare Biosciences). Mem-branes were blocked with 5% nonfat dry milk for 1 hour atroom temperature and incubated with appropriate primaryantibodies overnight at 4�C. Afterwashingwith PBS contain-ing 0.1% Tween 20, the membrane was incubated with ahorseradish peroxidase–conjugated secondary antibody at a1:5,000 dilution and the signal was detected with a chemi-luminescence reagent (GE Healthcare Biosciences). Banddensity was determined using the Image J (NIH) softwareprogram.

Terminal deoxynucleotidyl transferase–mediateddUTP nick end labeling assay

Urinary bladder cancer tissues from female Fischer-344rats, induced with the carcinogen 4-hydroxybutylnitrosa-min (OH-BBN) and treated with 400mg/kg naproxen for 5days, as well as untreated controls, were prepared for theterminal deoxynucleotidyl transferase–mediateddUTPnickend labeling (TUNEL) assay and immunofluorescencemicroscopy. Apoptosis was determined by using the Dead-End Colorimetric TUNEL System (Promega) according tothe manufacturer’s instructions. Briefly, after deparaffiniza-tion and rehydration, the tissue sections were pretreatedwith 20mg/mLproteinaseK solution for 10minutes at roomtemperature. Thereafter, slides were rinsed in PBS andincubated with TUNEL reaction mixture for 1 hour at37�C in a humidified chamber. Slides were then washedwith PBS followed by a stop solution for 10 minutes atroom temperature. The slides were counterstained with40,6-diamidino-2-phenylindole (DAPI) under glass cover-slips. The stained tissue was examined at �200 magnifica-tion using a Nikon Eclipse TE2000-E Confocal microscope.

Immunofluorescence microscopySlideswere baked at 60�C for 2hours, deparaffinizedwith

xylene, and rehydrated through a graded alcohol bath. Ofnote, 10 mmol/L sodium citrate buffer (pH 6.0) was used






for antigen retrieval by heating slides and buffer for 10minutes in the microwave followed by 20 minutes of ambi-ent cooling. Slides were washed with DI water followedby 1� PBS. For immunofluorescence, specimens wereblocked in 5% donkey serum–1� PBS–0.3% Triton bufferfor �1 hour. The primary antibody was prepared accordingto the manufacturer’s instructions and left on overnight at4�C. Slides were washed in 1� PBS followed by 2 hours ofappropriate secondary antibody incubation preparedaccording to the manufacturer’s instructions. Slides werewashed with 1� PBS followed by high-salt PBS (23.38 gNaCl in 1 L 1� PBS). Fluoro-Gel II with DAPI (ElectronMicroscopy Sciences, #17985-50) was used to mount glasscoverslips and then sealed using clear nail polish. All stainedtissues were examined at�200magnification using a NikonEclipse TE2000-E Confocal microscope.

Statistical analysisAll quantitative results are expressed as mean � SD.

Statistically significant differences were obtained using theStudent t test or by one-way ANOVA. A P < 0.05 value wasconsidered to be statistically significant.

ResultsVirtual screening for novel targets of naproxen

Shape similarity screening was used to search for thepotential kinase targets of naproxen. The screening wasperformed using the phase module of Schr€odinger-Maestrov9.2 to compare the volumetric similarity betweennaproxenand each compound from the PDB ligand database (12).Eight different potential kinase targets (column IV) fornaproxen were identified from shape screening (Table 1).The ligands (column I) from the crystal structures (columnIII) of these kinase targets have shape similarity scores of atleast 0.75 (column II) when compared with naproxen.

Naproxen inhibits PI3K activity directlyWe examined whether naproxen had an effect on the

activity of several kinases that were identified by our virtualscreening. We selected 5 kinases (PI3K, JNK1, Akt1, Akt2,and Src) based on their importance in cancer development

and commercial availability. Naproxen only dose depen-dently suppressed the kinase activity of PI3K (Fig. 1Band Supplementary Fig. S1). Therefore, PI3K is the mostlikely target for naproxen. Next, we investigated whethernaproxen interacted directly with PI3K. The direct bindingof naproxen and PI3K was demonstrated by an in vitro pull-down assay (Fig. 1C). In addition, we observed ex vivobinding between naproxen and PI3K in UM-UC-5 celllysates (Fig. 1D). To elucidate the potential binding site,we conducted an in silico docking study using the induced fitdocking module from Schr€odinger. The hierarchical dock-ing algorithm glide of Schr€odinger-Maestro v9.2 (13) wasthen used to assess the possible binding orientations ofnaproxen and PI3K. The simulation returned a positiveresult, with the theoretical naproxen/PI3K complex shownin Fig. 1E andF. These data support PI3K as a potential targetof naproxen.

Naproxen attenuates downstream PI3K signalingEGFR is known to be involved in bladder carcinogenesis

and COX-2 and Akt, among others, are all potential down-stream targets of EGFR and seem to be important moleculartargets of naproxen andotherNSAIDs (17,18). In this study,we used 2 cell lines with differing sensitivity to the EGFRinhibitor, gefitinib. Cell lines included UM-UC-5 (gefitinibsensitive) and UM-UC-14 (gefitinib resistant). To elucidatethe mechanism underlying the inhibitory effects ofnaproxen, we examined the effects of naproxen on themostwell-known downstreammolecules of PI3K, including Akt,mTOR, and p70S6K. Importantly, exposure of UM-UC-5(Fig. 2A) and UM-UC-14 (Fig. 2B) bladder cancer cells tonaproxen decreased the phosphorylation of Akt, mTOR,and p70S6K in both cell lines.

Naproxen decreases viability and inhibits anchorage-independent growth of UM-UC-5 and UM-UC-14urinary bladder cancer cells

Next, we determined whether naproxen could affectviability and anchorage-independent growth of UM-UC-5and UM-UC-14 bladder cancer cells. Naproxen was foundto significantly decrease viability of bothUM-UC-5 (Fig. 3A,

Figure 2. Naproxen suppressesphosphorylation of the PI3K/Aktsignaling pathway. A, cells weretreatedwith naproxen (0, 0.5, 1, or 2mmol/L) in medium containing10%FBSand analyzed byWesternblot analysis. Naproxen inhibitsphosphorylation of Akt,mTOR, andp70S6K in UM-UC-5 (A) and UM-UC-14 (B) cells. The bands werequantified by densitometricanalysis using Image J softwareand normalized using theirnonphospho-band.

Kim et al.





left) andUM-UC-14 (Fig. 3A, right) cells. We alsomeasuredanchorage-independent cell growth using a soft agar assayand naproxen inhibited growth of UM-UC-5 (Fig. 3B) andUM-UC-14 (Fig. 3C) cell in soft agar.

Naproxen induces G0–G1 cell-cycle arrest in UM-UC-5and UM-UC-14 urinary bladder cancer cellsTo elucidate the mechanism of naproxen’s inhibitory

effect on cell growth, we used flow cytometry analysis afterpropidium iodide staining. Previous reports show that thePI3K inhibitor, LY294002, induced G0–G1 phase arrest inmany types of cancer cells (19–21). Naproxen alsoincreased the percentage of cells in G0–G1 in both UM-UC-5 (Fig. 4A) and UM-UC-14 (Fig. 4B) cells after 48 hourstreatment and a corresponding decrease was shown in Sphase. LY294002 treatment was reported to arrest the cellcycle at G0–G1 through G1-associated proteins including

p21cip, CDK4, and cyclinD1 (19). Treatmentwithnaproxenfor 48 hours also increased p21cip and decreased CDK4 andcyclin D1 expression in both UM-UC-5 (Fig. 4C) and UM-UC-14 (Fig. 4D) cell lines.

Naproxen induces apoptosis in UM-UC-5 andUM-UC-14 urinary bladder cancer cells

The PI3K signaling pathway delivers an antiapoptoticsignal (22) and inhibition of PI3K activity induces apopto-sis (23, 24). We examined the effect of naproxen on apo-ptosis in UM-UC-5 and UM-UC-14 bladder cancer cellsusing Annexin V/propidium iodide double staining.Annexin Vonly stains cells representative of early apoptosis,which was increased in naproxen-treated UM-UC-5 (Fig.5A) and UM-UC-14 (Fig. 5B) cells. We then studied theeffect of naproxen on effectors of the cell death pathways.Naproxen inducedPARP, caspase-3, and caspase-7 cleavage.

Figure 3. Naproxen exertsanticancer activity against urinarybladder cancer cells. A, naproxeninhibits bladder cancer cell growthin a dose-dependent manner.UM-UC-5 (left) or UM-UC-14 (right)cells (1 � 103 cells/well) wereseeded in 96-well plates andtreated with variousconcentrations (0, 0.5, 1, or 2mmol/L) of naproxen andproliferation was measured byMTS assay. Data are shown asmeans � SD (N ¼ 5) and theasterisk (�) indicates a significant(P < 0.05) difference in naproxen-treated cells compared withuntreated control cells. Naproxeninhibits anchorage-independentcancer cell growth. UM-UC-5 (B) orUM-UC-14 (C) cells wereincubated in 0.3% agar for 10 dayswith various concentrations (0, 0.5,1, or 2 mmol/L) of naproxen.Colonies were counted using amicroscope and the Image-ProPLUS (v.6.1) computer softwareprogram. Representativephotographs are shown in the leftpanels and data in the right panelsare represented as means � SD ofvalues from triplicates and similarresults were obtained from 2independent experiments. Theasterisk (�) indicates a significant(P < 0.05) decrease in colonyformation induced by naproxencompared with untreated control.






The ratio of proapoptotic factor Bax and prosurvival factorBcl-2, which is important for apoptosis, and both proteinsassociated with the PI3K pathway were also examined. Wefound that naproxenmarkedly increased the Bax/Bcl-2 ratioin both UM-UC-5 (3.7/0.5–5/0; Fig. 5C) and UM-UC-14(1.8/0.7–1.5/0.2; Fig. 5D) cells.

Naproxen induced apoptosis and inhibited Aktphosphorylation in OH-BBN–induced bladder cancer

Our previous study indicated that naproxen inhibitsOH-BBN–induced bladder cancer in rats (6). Naproxenwas effective in decreasing the incidence of large tumors bygreater than 85%at sacrifice (weight of bladder plus lesions

Figure 4. Naproxen induces G1

arrest in UM-UC-5 (A) and UM-UC-14 (B) urinary bladder cancer cells.Cells were starved in serum-freemedium for 24 hours and thentreatedwith naproxen (0, 0.5, 1, or 2mmol/L) for 48 hours. Cell-cycleanalysis was performed using flowcytometry. Data are shown asmean percentages � SD (N ¼ 3)and the asterisk (�) indicates asignificant (P < 0.05) difference innaproxen-treated cells comparedwith untreated control cells.Naproxen regulates the CDK4/cyclin D1 signaling pathway in UM-UC-5 (C) and UM-UC-14 (D) cells.Cells were treated with naproxen(0, 0.5, 1, or 2 mmol/L) in 10%FBS/DMEM. The levels of E2F,cyclin D1, cyclin A, CDK2,CDK4, and p21cip proteins weredetermined by Western blotanalysis. The bands werequantified by densitometricanalysis using Image J softwareand normalized against b-actin.

Figure 5. Naproxen inducesapoptosis in UM-UC-5 (A) andUM-UC-14 (B) cells. Cells were starvedin serum-free medium for 24 hoursand then treated with naproxen(0, 0.5, 1, or 2mmol/L) for 72 hours.Apoptosis was measured by flowcytometry. Data are shown asmeans � SD (N ¼ 3) and theasterisk (�) indicates a significant(P < 0.05) difference in naproxen-treated cells compared withuntreated control cells. The levelsof cleaved caspase-3, caspase-7,PARP, and expression levels ofp53, Bcl-2, and Bax weredetermined in UM-UC-5 (C) andUM-UC-14 (D) cells by Westernblot analysis. The bands werequantified by densitometricanalysis using Image J softwareand normalized using b-actin.

Kim et al.





greater than 200 mg) even when treatment was initiatedwhen microcarcinomas were already present (6). Usingbladder tissue from short-term treatment with naproxen(Materials and Methods), we examined apoptosis byTUNEL and DAPI staining and Akt phosphorylation.Apoptosis was significantly increased by naproxen treat-ment in animal tissue samples (Fig. 6A and B). Further-more, Akt phosphorylation (Ser473) was decreased innaproxen-treated versus untreated urinary bladder tumors(Fig. 6C and D).

DiscussionMore than 70,000 new cases of urinary bladder cancer

were diagnosed in the United States in 2010, with 14,680deaths (25). Furthermore, bladder cancer treatment is themost expensive per patient comparedwith all other cancers.More than 90% of bladder cancers cases are urothelialcarcinomas. A number of groups have reported that variousNSAIDs and celecoxib are striking inhibitors of chemicallyinduced urinary bladder cancer in rodent animal models(7, 26, 26). Furthermore, the NSAID piroxicam has provento be effective in canine bladder cancer (8–10). Finally anumber of epidemiologic studies imply the efficacy ofNSAIDs in reducing human bladder cancer (11). However,

in much of the epidemiologic literature, this response hasappeared variable (25, 28–30).

The most serious side effects associated with mostNSAIDs include increased adverse cardiovascular eventsand gastrointestinal ulceration (4, 5). Naproxen is a well-knownNSAID that is easily purchasedover-the-counter andhas lower cardiovascular toxicities than any of the otherNSAIDs and is in fact recommended for persons witharthritis who have a higher cardiovascular risk. Similar tomost other NSAIDs and higher doses of aspirin, naproxencan cause gastric toxicity and bleeding. However, this effectcan apparently be amelioratedwithout losing efficacy eitherby combining it with omeprazole or altering the dosingschedule (R.A. Lubet and C.J. Grubbs; unpublished data).

Naproxen seems tobe a good agent for drug repositioningand might be a more useful drug if other applications werefound. To determine other applications, exploring themolecular mechanism and direct molecular target(s) ofnaproxen is critical. Several previous studies have shownthat naproxen has anti-urinary bladder cancer effects in vivo(6, 7). In this study, we identified PI3K as an additionalpotential molecular target for naproxen and elucidated itsmechanism of action through this protein target. The PI3Ksignaling pathway plays a key role in cancer cell growth,survival, motility, andmetabolism (31) and is activated in a

Figure 6. Naproxen inducesapoptosis and decreasesphosphorylation of Akt inmouse bladder cancertissues. Bladder cancer wasinduced in female Fischer-344rats with the carcinogen4-hydroxybutylnitrosamin(OH-BBN) and then rats weretreated or not treated with400 mg/kg of naproxen for 5 days.A, B, naproxen induces apoptosisin vivo. The TUNEL assay wasconducted following themanufacturer's protocol to detectapoptosis inmouse bladder cancertissues. C, D, naproxen inhibits Aktphosphorylation in vivo. Fixedtumor samples from rats in A weresubjected to immunofluorescencestaining to detect phosphorylatedAkt (Ser473) by incubating withspecific primary antibodies andsecondary antibodies conjugatedto a fluorescence dye. Sampleswere observed by confocalmicroscopy. For A and C,representative images are shownfrom a minimum of 6 tumors withsimilar results. For B and D, dataare shown as means � SD and theasterisks (� and ��) indicate asignificant (P < 0.05 and P < 0.01,respectively) difference innaproxen-treated bladder cancertissues compared with untreatedbladder cancer tissues.






variety of different human cancers. PI3K belongs to a familyof lipid kinases that phosphorylate the 30-hydroxy positionof the inositol ring of phosphatidylinositides, yieldingproducts of which the best characterized is PIP3, the secondmessenger that recruits Akt to the cell membrane andactivates Akt. Akt activation is associated with variousdownstream signals, which are involved in carcinogenesis(32). Inhibitors of this pathway are under active develop-ment as anticancer agents (33). We tested the inhibitoryeffects of naproxen on 5 kinases, including PI3K, JNK1,Akt1, Akt2, and Src in the target list of naproxen identifiedby our virtual screening. In this study, we first showed thatnaproxen binds with PI3K and inhibits its activity resultingin suppression of Akt and p70S6K signaling downstream ofPI3K in vitro and in vivo (Fig. 7). To investigate themolecularbasis of PI3K inhibition by naproxen, we performed adocking study using a homology model structure of PI3K.Naproxen could bind well at the ATP binding pocket ofPI3K. Several important hydrogen bonds were formedbetween naproxen and PI3K, including Val882 at the back-bone and Lys833 at the side chain of PI3K. Naproxen also

formed p–p stacking with Tyr867 and hydrophobic inter-actions with Met804, Pro810, Trp812, Ile831, Ile879,Ala885, Ile881, Met953, Phe961, and Ile963. This suggeststhat naproxen could be a potential inhibitor against PI3K.These results add a new perspective to our understanding ofthe parameters involved in PI3K inhibition by naproxen.Further studies using X-ray crystallography or nuclear mag-netic resonance techniques are needed to confirm the exactbinding mode of naproxen to PI3K. Other targets identifiedby virtual screening such as urokinase, a serine protease, andPKA–PKB chimera are interesting potential targets andsuggest a number of possible additional mechanism ofaction (34). These and other potential protein targets willbe the subject of further studies.

In summary, we have shown that naproxen decreasesviability and inhibits anchorage-independent growth ofurinary bladder cancer cells by binding and suppressingPI3K activity and downstream signaling. Inhibition of thesesignaling pathways resulted in cell-cycle arrest and apopto-sis. These studies demonstrate that naproxen can interactwith PI3K and that this interaction can contribute to theefficacy of naproxen and may offer some additional usefulbiomarkers.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: M.-S. Kim, J.-E. Kim, Z. Huang, R.A. Lubet,Z. Dong, A.M. BodeDevelopment of methodology: D.Y. Lim, C.J. GrubbsAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): M.-S. Kim, A. Langfald, C.J. GrubbsAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis):M.-S. Kim, D.Y. Lim, Z. Huang, H. Chen,A. Langfald, Z. Dong, A.M. BodeWriting, review, and/or revision of themanuscript:M.-S. Kim, J.-E. Kim,Z. Huang, H. Chen, A. Langfald, R.A. Lubet, A.M. BodeAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): Z. DongStudy supervision: Z. Dong, A.M. Bode

Grant SupportThis work was supported by The Hormel Foundation and National

Institutes of Health Contract HHSN-261200433009C and NO1-CN-55006-72.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

ReceivedAugust 6, 2013; revisedNovember 5, 2013; acceptedDecember 5,2013; published OnlineFirst December 10, 2013.

References1. Sleigh SH, Barton CL. Repurposing strategies for therapeutics. Pharm

Med 2010;24:151–9.2. Ashburn TT, Thor KB. Drug repositioning: identifying and developing

new uses for existing drugs. Nat Rev Drug Discov 2004;3:673–83.3. da Cunha EF, Ramalho TC, Josa D, Caetano MS, de Souza TC.

Targeting inhibition of COX-2: a review of patents, 2002–2006. RecentPat Inflamm Allergy Drug Discov 2007;1:108–23.

4. Avorn J. Keeping science on top in drug evaluation. N Engl J Med2007;357:633–5.

5. Trelle S,ReichenbachS,Wandel S,HildebrandP, TschannenB, VilligerPM, et al. Cardiovascular safety of non-steroidal anti-inflammatorydrugs: network meta-analysis. BMJ 2011;342:c7086.

6. Lubet RA, Steele VE, Juliana MM, Grubbs CJ. Screening agents forpreventive efficacy in a bladder cancer model: study design, end

Figure 7. A scheme for illustrating the effect of naproxen on the regulationof cell cycle and apoptosis mediated through the inhibition of the PI3Ksignaling pathway.

Kim et al.





points, and gefitinib and naproxen efficacy. J Urol 2010;183:1598–603.

7. Steele VE, Rao CV, Zhang Y, Patlolla J, Boring D, Kopelovich L, et al.Chemopreventive efficacy of naproxen and nitric oxide-naproxen inrodent models of colon, urinary bladder, and mammary cancers.Cancer Prev Res 2009;2:951–6.

8. HenryCJ,McCawDL, Turnquist SE, Tyler JW,BravoL, Sheafor S, et al.Clinical evaluation of mitoxantrone and piroxicam in a canine model ofhuman invasive urinary bladder carcinoma. Clin Cancer Res2003;9:906–11.

9. Mohammed SI, Bennett PF, Craig BA, Glickman NW, Mutsaers AJ,Snyder PW, et al. Effects of the cyclooxygenase inhibitor, pirox-icam, on tumor response, apoptosis, and angiogenesis in a caninemodel of human invasive urinary bladder cancer. Cancer Res2002;62:356–8.

10. Robat C, Burton J, Thamm D, Vail D. Retrospective evaluation ofdoxorubicin-piroxicam combination for the treatment of transitionalcell carcinoma in dogs. J Small Anim Pract 2013;54:67–74.

11. BarisD, KaragasMR, Koutros S,Colt JS, JohnsonA, SchwennM, et al.Nonsteroidal anti-inflammatory drugs and other analgesic use andbladder cancer in northern New England. Int J Cancer 2013;132:162–73.

12. Lejon S, Cramer JF, Nordberg P. Structural basis for the binding ofnaproxen to human serum albumin in the presence of fatty acids andthe GA module. Acta Crystallogr Sect F Struct Biol Cryst Commun2008;64:64–9.

13. Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz DT,et al. Glide: a new approach for rapid, accurate docking and scoring. 1.Method and assessment of docking accuracy. J Med Chem 2004;47:1739–49.

14. Lubet RA, Zhang Z, Wiseman RW, You M. Use of p53 transgenic micein the development of cancer models for multiple purposes. Exp LungRes 2000;26:581–93.

15. Lubet RA, Lu Y, Bode AM, YouM, Verney ZM, Steele VE, et al. Efficacyof the EGFR inhibitor Iressa on development of chemically-inducedurinary bladder cancers: dose dependency andmodulation of biomar-kers. Oncol Rep 2011;25:1389–97.

16. Kwon JY, Lee KW, Kim JE, Jung SK, Kang NJ, Hwang MK, et al.Delphinidin suppresses ultraviolet B-induced cyclooxygenases-2expression through inhibition of MAPKK4 and PI-3 kinase. Carcino-genesis 2009;30:1932–40.

17. Grivas PD, Day M, Hussain M. Urothelial carcinomas: a focus onhumanepidermal receptors signaling. AmJTranslRes 2011;3:362–73.

18. Lu Y, Liu P, Van den Bergh F, Zellmer V, James M, Wen W, et al.Modulation of gene expression and cell-cycle signaling pathways bythe EGFR inhibitor gefitinib (Iressa) in rat urinary bladder cancer.Cancer Prev Res 2012;5:248–59.

19. Gao N, Zhang Z, Jiang BH, Shi X. Role of PI3K/AKT/mTOR signaling inthe cell cycle progression of humanprostate cancer. BiochemBiophysRes Commun 2003;310:1124–32.

20. Guillard S, Clarke PA, Te Poele R, Mohri Z, Bjerke L, Valenti M, et al.Molecular pharmacology of phosphatidylinositol 3-kinase inhibition inhuman glioma. Cell Cycle 2009;8:443–53.

21. Pene F, Claessens YE, Muller O, Vigui�e F, Mayeux P, Dreyfus F, et al.Role of the phosphatidylinositol 3-kinase/Akt and mTOR/P70S6-kinase pathways in the proliferation and apoptosis in multiple myelo-ma. Oncogene 2002;21:6587–97.

22. Kennedy SG, Wagner AJ, Conzen SD, Jordan J, Bellacosa A, TsichlisPN, et al. The PI 3-kinase/Akt signaling pathway delivers an anti-apoptotic signal. Genes Dev 1997;11:701–13.

23. Jiang H, Fan D, Zhou G, Li X, Deng H. Phosphatidylinositol 3-kinaseinhibitor(LY294002) induces apoptosis of human nasopharyngeal car-cinoma in vitro and in vivo. J Exp Clin Cancer Res 2010;29:34.

24. Shayesteh L, Lu Y, Kuo WL, Baldocchi R, Godfrey T, Collins C, et al.PlK3CA is implicated as an oncogene in ovarian cancer. Nat Genet1999;21:99–102.

25. Genkinger JM, De Vivo I, Stampfer MJ, Giovannucci E, Michaud DS.Nonsteroidal antiinflammatory drug use and risk of bladder cancerin the health professionals follow-up study. Int J Cancer 2007;120:2221–5.

26. Droller MJ, Gomolka D. Indomethacin and poly I:C in the inhibition ofcarcinogen-induced bladder cancer in an experimental animal model.J Urol 1984;131:1212–7.

27. Grubbs CJ, Lubet RA, Koki AT, Leahy KM, Masferrer JL, Steele VE,et al. Celecoxib inhibits N-butyl-N-(4-hydroxybutyl)-nitrosamine-induced urinary bladder cancers in male B6D2F1 mice and femaleFischer-344 rats. Cancer Res 2000;60:5599–602.

28. Daugherty SE, Pfeiffer RM, Sigurdson AJ, Hayes RB, Leitzmann M,Schatzkin A, et al. Nonsteroidal antiinflammatory drugs and bladdercancer: a pooled analysis. Am J Epidemiol 2011;173:721–30.

29. DovediSJ,DaviesBR.Emerging targeted therapies for bladder cancer:a disease waiting for a drug. Cancer Metastasis Rev 2009;28:355–67.

30. Leppert JT, Shvarts O, Kawaoka K, Lieberman R, Belldegrun AS,Pantuck AJ. Prevention of bladder cancer: a review. Eur Urol 2006;49:226–34.

31. Wong KK, Engelman JA, Cantley LC. Targeting the PI3K signalingpathway in cancer. Curr Opin Genet Devel 2010;20:87–90.

32. Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase-AKT path-way in human cancer. Nat Rev Cancer 2002;2:489–501.

33. Courtney KD, Corcoran RB, Engelman JA. The PI3K pathway as drugtarget in human cancer. J Clin Oncol 2010;28:1075–83.

34. Hildenbrand R, Gandhari M, Stroebel P, Marx A, Allgayer H, Arens N.The urokinase-system—role of cell proliferation and apoptosis. HistolHistopathol 2008;23:227–36.






2014;7:236-245. Published OnlineFirst December 10, 2013.Cancer Prev Res Mi-Sung Kim, Jong-Eun Kim, Do Young Lim, et al. by Targeting PI3K

CancersUrinary Bladder Cancer Cell Lines and Chemically Induced Naproxen Induces Cell-Cycle Arrest and Apoptosis in Human

Updated version

10.1158/1940-6207.CAPR-13-0288doi:

Access the most recent version of this article at:

Material

Supplementary

1

http://cancerpreventionresearch.aacrjournals.org/content/suppl/2013/12/11/1940-6207.CAPR-13-0288.DCAccess the most recent supplemental material at:

Cited articles

http://cancerpreventionresearch.aacrjournals.org/content/7/2/236.full#ref-list-1

This article cites 34 articles, 8 of which you can access for free at:

Citing articles

http://cancerpreventionresearch.aacrjournals.org/content/7/2/236.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerpreventionresearch.aacrjournals.org/content/7/2/236To request permission to re-use all or part of this article, use this link



http://cancerpreventionresearch.aacrjournals.org/lookup/doi/10.1158/1940-6207.CAPR-13-0288

http://cancerpreventionresearch.aacrjournals.org/content/suppl/2013/12/11/1940-6207.CAPR-13-0288.DC1

http://cancerpreventionresearch.aacrjournals.org/content/suppl/2013/12/11/1940-6207.CAPR-13-0288.DC1

http://cancerpreventionresearch.aacrjournals.org/content/7/2/236.full#ref-list-1

http://cancerpreventionresearch.aacrjournals.org/content/7/2/236.full#related-urls

http://cancerpreventionresearch.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerpreventionresearch.aacrjournals.org/content/7/2/236


Documents

NaproxenInducesCell-CycleArrestandApoptosisinHuman …...NaproxenInducesCell-CycleArrestandApoptosisinHuman Urinary Bladder Cancer Cell Lines and Chemically Induced Cancers by Targeting